Penerapan Machine Learning Operations Pada Sistem Klasifikasi Citra Fundus Dan Citra X-ray

Panjaitan, Hendrich Ardthian Breshman (2026) Penerapan Machine Learning Operations Pada Sistem Klasifikasi Citra Fundus Dan Citra X-ray. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5024221023-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (12MB) | Request a copy

Abstract

Citra fundus dan citra x-ray adalah citra medis yang memerlukan sistem analisis yang akurat dan andal. Deep learning, khususnya Convolutional Neural Network (CNN), telah banyak digunakan untuk klasifikasi citra fundus dan citra x-ray, namun penerapannya di lingkungan nyata masih menghadapi tantangan dalam hal otomatisasi, Training, dan monitoring sistem. Oleh karena itu, diperlukan pendekatan Machine Learning Operations (MLOps)untuk mengelola siklus hidup model secara end-to-end. Penelitian ini mengimplementasikan pipeline MLOps untuk sistem klasifikasi citra fundus dan citra x-ray dengan memanfaatkan n8n sebagai workflow automation tool. Pipeline yang dibangun mencakup tahapan generate dataset, pelatihan model,retraining, serta monitoring performa sistem. Model klasifikasi dikembangkan menggunakan arsitektur CNN. Selain itu, metode Explainable Artificial Intelligence (XAI) berupa Gradient-weighted Class Activation Mapping (Grad-CAM) diterapkan untuk memberikan visualisasi area citra yang berkontribusi terhadap hasil prediksi model, sehingga meningkatkan interpretabilitas dan transparansi sistem. Hasil penelitian menunjukkan bahwa pipeline MLOps berbasis n8n mampu meningkatkan efisiensi dan keandalan proses serta evaluasi sistem klasifikasi citra fundus dan citra x-ray secara terintegrasi.
================================================================================================================================
Fundus images and x-ray images are medical images that require accurate and reliable analysis systems. Deep learning, particularly Convolutional Neural Networks (CNNs), has been widely used for fundus images and x-ray images classification; however, its application in real-world environments still faces challenges regarding automation, Training, and system monitoring. Therefore, a Machine Learning Operations (MLOps) approach is needed to manage the end-to-end model lifecycle. Study implements an MLOps pipeline for fundus images and x-ray images classification system using n8n as a workflow automation tool. The pipeline includes the stages of dataset generation, model training, retraining, and system performance monitoring. The classification model was developed using a CNN architecture. Additionally, the Explainable Artificial Intelligence (XAI) method known as Gradientweighted Class Activation Mapping (Grad-CAM) method is applied to visualize the image regions that contribute to the model’s prediction results, thereby enhancing the system’s interpretability and transparency. The results show that the n8n-based MLOps pipeline is capable of improving the efficiency and reliability of the process as well as the integrated evaluation of the fundus images and x-ray images classification system

Item Type:	Thesis (Other)
Uncontrolled Keywords:	MLOps, n8n, CNN, Grad-CAM, Citra fundus, citra x-ray. MLOps, n8n, CNN, Grad-CAM, fundus images, x-ray images.
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > Q Science (General) > Q337.5 Pattern recognition systems Q Science > QA Mathematics > QA336 Artificial Intelligence T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing T Technology > T Technology (General) > T58.62 Decision support systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User:	Hendrich Ardthian Breshman Panjaitan
Date Deposited:	24 Jul 2026 00:40
Last Modified:	24 Jul 2026 00:40
URI:	http://repository.its.ac.id/id/eprint/136617

Available Versions of this Item

• Penerapan Machine Learning Operations Pada Sistem Klasifikasi Citra Fundus Dan Citra X-ray. (deposited 24 Jul 2026 00:40) [Currently Displayed]
  • Penerapan Machine Learning Operations Pada Sistem Klasifikasi Citra Fundus Dan Citra X-ray. (deposited 24 Jul 2026 04:00)

Actions (login required)

View Item